Exploratory Learning

By Bhavana Dalvi, William W. Cohen, Jamie Callan. 2013

47 citations recorded by the source corpus.

In multiclass semi-supervised learning (SSL), it is sometimes the case that the number of classes present in the data is not known, and hence no labeled examples are provided for some classes. In this paper we present variants of well-known semi-supervised multiclass learning methods that are robust when the data contains an unknown number of classes. In particular, we present an "exploratory" extension of expectation-maximization (EM) that explores different numbers of classes while learning. "Exploratory" SSL greatly improves performance on three datasets in terms of F1 on the classes with seed examples—i.e., the classes which are expected to be in the data. Our Exploratory EM algorithm also outperforms a SSL method based non-parametric Bayesian clustering.

Subfields: Artificial Intelligence

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Lei Shu
2. Hu Xu
3. B. Liu
4. Bing Liu
5. Shuai Wang
6. Geli Fei
7. Zixuan Ke
8. William W. Cohen

Citations in the corpus

• DOC: Deep Open Classification of Text Documents (2017, methodology): “Dalvi et al. (2013) proposed a multi-class semi-supervised method based on the EM algorithm.”
• Learning Cumulatively to Become More Knowledgeable (2016, methodology): “In [11], three well-known multiclass semi-supervised learning methods were extended under the exploratory EM framework.”
• Breaking the Closed World Assumption in Text Classification (2016, methodology): “In (Dalvi et al., 2013), three well-known multiclass semi-supervised learning methods were extended under the exploratory EM framework.”
• Unseen Class Discovery in Open-world Classification (2018, background): “Several prior works exist on open classification that can reject unseen class examples (Scheirer et al., 2013; Jain et al., 2014; Fei & Liu, 2016; Scheirer et al., 2014; Jain et al., 2014; Dalvi et al., 2013).”
• Hierarchical Semi-supervised Classification with Incomplete Class Hierarchies (2016, methodology): “Our earlier work [12] had proposed similar approach in the context of flat classification (or clustering) setting, i.e., no hierarchical constraints between classes.”